# Posts on X before Sep 13, 2025, 6:36:12 AM UTC Canonical: https://michaelstolarz.com/x/before/1757745372000-1966752759085944871/ ## On X ### [Sep 10, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1965926505139351658) > When people have nothing to live for, when they are told that it's not their fault but someone else's, when their agency is algorithmically stripped away from their being, all that's left is hope for martyrdom in their echo chambered community. > > Rest in peace, Charlie Kirk. πŸ‡ΊπŸ‡ΈπŸ† Quoted post by Will Chamberlain: ### [Sep 10, 2025 β€” @willchamberlain](https://x.com/willchamberlain/status/1965862869968498877) > No one is going to do these events anymore. Town halls are done. One major political party has a subset of adherents that are perfectly happy to assassinate their political opponents if given the opportunity [x.com/AGHamilton29/s…]() ### [Sep 10, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1965716412829405434) > Same problem affected me; thankfully, almost done with the rewrite. Especially with so many endpoints to support... Quoted post by dax: ### [Sep 9, 2025 β€” @thdxr](https://x.com/thdxr/status/1965476829701267826) > opencode has a bunch of performance problems > > the reason we haven't shipped much in the last 2 weeks is we're fixing these by moving to opentui > > hoping to be back at parity by end of week ### [Sep 9, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1965424973037928730) > BIG UPDATE TODAY for my systemsculpt lovers. ### [Sep 6, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1964126444495523967) > MOM GET THE CLI > NEW [@openrouter]() STEALTH MODEL JUST DROPPED [Attached image](/x/media/f3e1ac60c0dd45575eeda8a6f10991293460e88ca5fe17dd57456b06c191b975.jpg) ### [Sep 4, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1963459876996223266) > once an llm provider figures out how to do per-account memory, even if it's just an intricate system vs. an llm architecture upgrade, it's game over (specifically in the coding space) > > imagine chatgpt's memory and how it utilizes previous chats so intelligently, PUT THAT IN CODEX ### [Aug 30, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1961793438405493080) > I mean, I don't get it. How do you not catch these? [@nikitabier]() surely there's a pattern to recognize here? [Attached image](/x/media/3946617e30f0a1c72c881a9ec0bba4aa5927c6dc83bffa455ea5003e71b4b2c6.png) ### [Aug 30, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1961769028579688653) > there are still people who think opus (or even sonnet) are better than gpt-5-high > > imagine\! ### [Aug 30, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1961743310306910505) > this is solved by having good taste and only ever staying in the Following tab, never For You Quoted post by kache: ### [Aug 30, 2025 β€” @yacineMTB](https://x.com/yacineMTB/status/1961740654028947533) > 50% of the timeline is chatgpt generated. \*Close app\* ### [Aug 26, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1960254012030337378) > hilariously ironic given this was supposed to be the hero against google > > ironic especially because you'll see more ads on firefox than on google, lmao Quoted post by near: ### [Aug 25, 2025 β€” @nearcyan](https://x.com/nearcyan/status/1960090227265224747) > what a new mozilla firefox install looks like in 2025. > > even the weather is 'sponsored content'. [Attached image](/x/media/667372232d63bbefbeae052a9989fd7dd4c9443731a3b4bf957827eddbebdcf3.jpg) ### [Aug 25, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1959969266444235080) > [patreon.com/posts/final-ho…]() ### [Aug 24, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1959556575544451368) > looking back, it's so cringe they didn't release claude code as open source. destined to fail from the start > > imagine sending out dmca's for a tui? lmao ### [Aug 23, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1959236229943230747) > wait until the MCP crowd finds out about CLI commands ### [Aug 23, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1959094106019303795) > If Codex CLI Pro is comparable to Claude Code Max, there’s no reason at all to use Claude Code Max. > > Features are easy to catch up to. Model intelligence and efficacy isn’t. Quoted post by Alexander Embiricos: ### [Aug 23, 2025 β€” @embirico](https://x.com/embirico/status/1959057942445269141) > Just bumped Codex CLI Plus limits up 50% as an experiment. Enjoy\! > > More transparency coming next week as things settle. [x.com/embirico/statu…]() ### [Aug 20, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1958052379586359521) > make sure to try out cursor's new sonic model O\_O [Attached image](/x/media/22b1bbace8fb0a39fedd5a1ccbff1cc3b1fd912056c89580af0cbcd53d2f710a.jpg) ### [Aug 19, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1957774766598676542) > few more years til you can have a 1 on 1 sparring buddy 24/7 for infinite workouts and sparring [Quoted post unavailable](https://x.com/i/status/1957474381216841994) ### [Aug 17, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1957038873273643297) > I hate gpt 5 high so much > > because it has made every other LLM unusable > > it never lies to me, > never makes the wrong decision, > tells me when it doesn't know/needs more info. > > damn. thanks a lot openai. ### [Aug 16, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1956694396185010328) > pls ser i take back what i said about u anthropic pls i have 1 children to feed with my plugin and it needs updates pls πŸ₯Ί [Attached image](/x/media/0b57cdfb60a855767cb783cae531d9ad1be75ddc9ff64b96e785ba82dccfdff5.jpg) ### [Aug 16, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1956631781693637110) > this is why anthropic will lose and openai/google will win > > the only plus they've had so far is first mover to tool call > > now ironically because of all the safety stuff they've put on top of their models, > > the models lie, cheat, and kill themselves. yes, really. Quoted post by Anthropic: ### [Aug 15, 2025 β€” @AnthropicAI](https://x.com/AnthropicAI/status/1956441209964310583) > As part of our exploratory work on potential model welfare, we recently gave Claude Opus 4 and 4.1 the ability to end a rare subset of conversations on [claude.ai](). [Attached image](/x/media/b712706eaf3e7c1d8a4da66ba4f0b3cfcfa098e958954cabe2637ceb0ea38702.png) ### [Aug 16, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1956533688365187436) > this is the biggest difference between anthropic and openai models > > anthropic models talk sweet to you, they trick you that they'll do all the work well, they lie, cheat on tests, they make mock data > > openai models have caught up - and they are NOT lazy. they are EXCITED to work. [Attached image](/x/media/4ee3a9230dc19a260586cc06585881e00fee3fc043d1030908f368672698d2c5.jpg) ### [Aug 15, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1956388063170748923) > so weird to see this in gpt 5's thinking traces > thankfully it keeps going? [Attached image](/x/media/9c89f4fb9997b00ea90034dd3e8ba31ca6cb4c4500b960af7d7e9bcb3ecbe18d.jpg) ### [Aug 15, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1956371121802215815) > new v2.0.0 of systemsculpt out > > lots of stability updates, byok, ease of use, standard/advanced modes, yeah lots of cool stuff > > more official release info later today\! [Attached image](/x/media/741dd3ea773e4a1d92b40e9cf42c31e56733b79e66234fa07004ff8f469c7eca.jpg) ### [Aug 15, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1956310141424271562) > so cool. can someone show me the usa equivalent? no? why not? why does china have 40 types of robots from 20 different startups and we have 3 can-barely-walk-but-marketed-so-well slop bots? Quoted post by CN Wire: ### [Aug 15, 2025 β€” @Sino_Market](https://x.com/Sino_Market/status/1956246931480641718) > πŸ‡¨πŸ‡³The 100-meter dash at the World Humanoid Robot Games, Beijing, China > [\#CHINA]() [\#ROBOTICS]() [\#humanoid]() [\#HumanoidRobots]() [Video on X](/x/media/fa41df2bca1deeed81e6a90a8ef06140c843cfeaa2dbc7dd2258eff010a0f647.jpg) ### [Aug 15, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1956145760426451017) > if i dont finish this automated commit + release system for systemsculpt today im gonna whip myself like silas from the da vinci code [Attached image](/x/media/5df8df2b37a30bfa4b95599c699159549c260a2a339ce72f193267034911eee6.jpg) ### [Aug 14, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1955948122833817611) > i hate ironing out the final bugs > they're final bugs for a reason ### [Aug 13, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1955734757473706151) > this is how much time there's left to use gpt 5 high fast on cursor for free btw [Attached image](/x/media/fbd0444730bab332c7c8f8bf2ec1bbeabb8a906414a7ed7162a4a9b62ae8ef66.jpg) ### [Aug 13, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1955731055891374518) > i mean this is the type of shit you'd read schitzos post about on /pol > > and yet here we are lmao Quoted post by James Harvey: ### [Aug 12, 2025 β€” @JamesHarvey2503](https://x.com/JamesHarvey2503/status/1955215331959394764) > 🚨 Man given SIX POINTS on his driving license, BEFORE he even has one (which is an automatic ban), because he used a SKATEBOARD WITHOUT INSURANCE. > > This country is an absolute joke. [Video on X](/x/media/4a92f453165a7033542e1ca944902bd38923482d14c2550dfd63721a72907129.jpg) ### [Aug 13, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1955468808669610010) > lets take sonnet 4 1m for a spin [Attached image](/x/media/2f3cf217b6f28b62d9aa1159efbd741cc4e84f4640185de32d96df2d980bae07.jpg) ### [Aug 13, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1955447386962268458) > how did they greenlight this for release [Attached image](/x/media/8c2a66f996d12aa4d7d311aecc9389b194828f086d26d5cc8074e8b65adc9c27.jpg) ### [Aug 12, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1955284840829309151) > i just know sama is gonna get banned soon > that or elon is about to go demon mode in everything else [Attached image](/x/media/6bbf6373f9bba78300926d873f9a403ca8cd5aa6db2311b7087ccaaa705cb213.jpg) ### [Aug 12, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1955261983806173695) > what have they done to my boy chatgpt, omg [Attached image](/x/media/88784ef2b76748951fb2b621a7eb813fdd4cfe86cd9e9a1282c9dc715808de5a.jpg) ### [Aug 12, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1955227897578627482) > waiting for this > until then im using gpt 5 high Quoted post by Anthropic: ### [Aug 5, 2025 β€” @AnthropicAI](https://x.com/AnthropicAI/status/1952768435873612256) > We plan to release substantially larger improvements to our models in the coming weeks. ### [Aug 12, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1955115857753018738) > the reason that gpt 5 high (specifically talking about coding here) is the current SotA, is that it doesn't lie > > aka doesn't hallucinate, or at least nowhere near as much as ex. opus 4.1 or sonnet 4 > > the simple fact of lie % is what pushes its efficacy beyond the horizon ### [Aug 11, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1954876788045476303) > why is chatgpt practically useless now? every answer it gives is trash, made for speed of response instead of quality of response, even when you run it in thinking mode it takes 10-15 seconds at most before giving you a slightly better slop answer, wtf openai ### [Aug 9, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1954027485013979295) > this memory update i'm cooking up for systemsculpt... it's like nothing anyone has experienced before > > not claude code, not gpt 5, not cursor. there's a reason why i've studied the blade in silence for these past few months with little to no interaction with others ### [Aug 8, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1953696306536337876) > chatgpt's 4o finding out it's joever > being replaced by 5 > > good night sweet prince [Attached image](/x/media/365da18697a9e34d9482316b38af221b578ff90adc6ead99221f896137d03c06.jpg) ### [Aug 8, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1953681636643815645) > first a brain dump, then my top 3: > > for it to be more of a pleasure to spend hours on end within. concise vs. verbose tool call output mode. planning mode (or commands with ability to constrain tools ex. no editing, only researching). command timeouts (current one 800s deep)... [Quoted post unavailable](https://x.com/i/status/1953590703453614386) ### [Aug 7, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1953559485056008505) > OH MY GOD Quoted post by Cursor: ### [Aug 7, 2025 β€” @cursor_ai](https://x.com/cursor_ai/status/1953559384531050724) > Cursor is now in your terminal\! It’s an early beta. > > Access all models. Move easily between your CLI and editor. [Attached image](/x/media/92b6cd8acc6c4439d6d469ce20cb67164fb5cfe286b70e5ea82569a56189b2e6.jpg) ### [Aug 7, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1953557679475241448) > i just dont get it, why is the open source claude code killer codex still absolute garbage when it comes to tui? > > [@OpenAIDevs]() let me show you a few examples, a 🧡: [Attached image](/x/media/8b51d4cd90426b56d7a1b7dc87ad0b6727205f910102ec4564cec3314ba923eb.jpg) ### [Aug 7, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1953548384268349786) > thank God they fixed naming though that's a relief, it's so intuitive as to which to choose, i love it [Attached image](/x/media/ed518b56d267aa47a4e84644879b4a0438def8a3f01e9e24219ed85b08147972.jpg) ### [Aug 7, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1953395736667664663) > this sounds scary > i first felt this with o3 in cursor > it would read 30 files, make 3 edits, then say ok im done > > i have a feeling it will be like that but on steroids Quoted post by sankalp: ### [Aug 7, 2025 β€” @dejavucoder](https://x.com/dejavucoder/status/1953362017332482400) > hyped for gpt5. also scared a bit. [x.com/deedydas/statu…]() ### [Aug 7, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1953286429951324509) > will it be unlimited use with the $200 plan? please? Quoted post by Dylan Hunn: ### [Aug 7, 2025 β€” @dylhunn](https://x.com/dylhunn/status/1953283427324838001) > no better time to npm i -g @openai/codex ### [Aug 5, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1952782581566718022) > these prices are mind boggling > > this is basically: > - o3-level intelligence > - flash-level pricing/speed [Attached image](/x/media/7c7ab3ea17321fc3d14dbc42166d1a19fac56fbaef4da9af81569b2493594fd8.jpg) ### [Aug 5, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1952774848532693153) > btw if you're trying to use the new opus 4.1 dont forget to input: > > /model claude-opus-4-1-20250805 > > at least for me, when i do /model opus right now, it still uses old 4.0 [Attached image](/x/media/28c84490a0aaa7b89cd6eb8904089e2962875ab25e62e4969b6cb9ec431ab773.jpg) ### [Aug 5, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1952773153106927791) > with opus 4.1 there's now no good reason to use sonnet 4 (given we're talking about max20 claude code plan here) > > it used to be that sonnet 4 could potentially outperform especially with speed opus 4, but seems like opus 4.1 is across the board better now [Attached image](/x/media/91a002be6ff5f9af28ad99409d8e858691e6276d323e75049460ab468c0593e1.jpg) ### [Aug 4, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1952214855689392328) > keep your memories clean > store everything as markdown/plaintext > this is a short term victory but a victory nonetheless > > in 5ish years though? it wont matter > you'll just swarm xyz application you used to use, > it will consume every bit of data it possibly can > and migrate you easily [Attached image](/x/media/46bb6f4b235bb6e2fe1095f47abfcd4e454af609c32e8dc07ae47a0c6b60cb43.jpg) ### [Aug 3, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1952078803288969383) > you can now live in the cli btw and do everything from there > > i mean you always could, but now you can make fully custom fine tuned golang tview interfaces perfected for your exact daily workflow... in about a few hours worth of tinkering [Attached image](/x/media/9e2fb4af4adb600a8f061b04b4aa54bef5c01f3b67ddac06bd98d9fba779d1a8.jpg) ### [Aug 2, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1951609238083809609) > while i agree, the problem is > until openai catches up with... > > 1. a coding model surpassing unlimited Opus 4 > 2. a CLI 2x better than Claude Code (codex is TRASH) > 3. aforementioned unlimited coding model included in their already-existing $200 plan > > ...we're stuck with Daddy Dario Quoted post by Ahmad: ### [Aug 2, 2025 β€” @TheAhmadOsman](https://x.com/TheAhmadOsman/status/1951477677526835665) Β· Edited > some of anthropic rugpulls so far: > > \> windsurf no access to claude 4 > > \> plus not getting opus 4 in claude code > > \> 1.58-bit quantized models during daytime > > \> max plans limits cut in half 2 weeks ago, no comms > > \> weekly limits without concrete numbers > > \> 5x/20x plans being actually 3x/8x of plus > > \> cutting off openai api access > > what an absolutely horrible company ### [Aug 2, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1951584027447857311) > this is an amazing example > > of thinking that you're being super productive and making use of your max 200 claude code plan > > the reality is this is much, much slower than one instance per project > > regarding the straight to main and skip perms - that's totally fine, i do that too [Attached image](/x/media/0686b7be88100b464c23c8d4cff60ebf8186caf46ce2cc3c196115130ba52454.jpg) ### [Aug 2, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1951554184421187879) > unfortunately they are still dogwater at coding > > yeah they can use a few tools and one shot things pretty ok but they totally fall apart further with every follow up prompt you make to it Quoted post by Chris: ### [Aug 2, 2025 β€” @ChrisGPT](https://x.com/ChrisGPT/status/1951475909099233448) > Horizon Beta - Xbox 360 controller. > > The models are definitely getting better [Attached image](/x/media/e4364e76cd1bc05cc1b7743ab12fd61b4e10a2076df6db4abdcd36acde2e40d7.jpg) ### [Aug 2, 2025 β€” @MichaelStolarz](https://x.com/MichaelStolarz/status/1951450585930330217) > Welp, time to do another all nighter and see what YET ANOTHER [@openrouter]() stealth model is capable of. [Attached image](/x/media/75322a2cd9d07cfd292755df95c5badfdfce425adb1df7d60cf42190b7e65d7f.png) [Older posts](/x/before/1754097050000-1951450585930330217/index.md) [More from @michaelstolarz](https://x.com/michaelstolarz) Saved from X: 2026-09-07T03:44:39.289Z